def predict_vm(ecs_infor_array, input_file_array):
#Get the CPU information
CPU_kernal, CPU_memory, N_Pvm, condition, Pvm, Predict_time = utils.splitEscData(
ecs_infor_array)
'''
CPU_kernal: The number of CPU kernal
CPU_memory: The size of memory
N_Pvm: The number of Vms which need to predict
condition
Pvm: The Vms which need to predict
Predict_time: The period of prediction
'''
#Get the History Data information
length, Hvm = utils.splitInputData(input_file_array)
#History Data
#Statistic and Split
lenD, S_Data = utils.Statistic_Split(length, Hvm, N_Pvm, Pvm)
result = []
if ecs_infor_array is None:
print 'ecs information is none'
return result
if input_file_array is None:
print 'input file information is none'
return result
#---------------------method one----------------------
#NEPvm=lwlr.Prob_predict(N_Pvm,lenD,S_Data,Predict_time)
#---------------------method two----------------------
NEPvm = lwlr.lwlr_predict(N_Pvm, lenD, S_Data, Predict_time)
print NEPvm
#write result
result.append(int(sum(NEPvm)))
for i in range(N_Pvm):
result.append((str(Pvm[i][0]) + " " + str(int(NEPvm[i]))))
result.append("")
# the numbor of predicted vms
total_kernal = 0
total_memory = 0
for i in range(N_Pvm):
total_kernal = total_kernal + NEPvm[i] * Pvm[i][1]
total_memory = total_memory + NEPvm[i] * Pvm[i][2]
if (math.ceil(total_kernal / CPU_kernal) >= math.ceil(
total_memory / CPU_memory)):
N_PCPU = int(math.ceil(total_kernal / CPU_kernal))
else:
N_PCPU = int(math.ceil(total_momery / CPU_kernal))
# allocation
CPU, N_PCPU = Boxing(N_PCPU, NEPvm, Pvm, N_Pvm, CPU_kernal, CPU_memory)
result = utils.results_expression(result, CPU, N_PCPU)
return result
def predict_vm(ecs_infor_array,input_file_array):
#Get the CPU information
N_Stype,Servers,N_Pvm,Pvm,Predict_time,Predict_Start=utils.splitEscData(ecs_infor_array)
#Get the History Data information
length,Hvm,History_End=utils.splitInputData(input_file_array)
Hdate=datetime.datetime.strptime(History_End,"%Y-%m-%d %H:%M:%S").date()
#Statistic and Split
lenD,S_Data=utils.Statistic_Split(length,Hvm,N_Pvm,Pvm)
# deal with gaps
gaps=int(round((utils.getTimeDiff(Predict_Start,History_End))/86400))
for i in range(N_Pvm):
for j in range(gaps):
S_Data[i].append((S_Data[i][-7]))
result = []
if ecs_infor_array is None:
print 'ecs information is none'
return result
if input_file_array is None:
print 'input file information is none'
return result
if(N_Pvm>5):
T=2
V=0
#return
NEPvm=ES.ES(lenD,N_Pvm,S_Data,Predict_time,T,V)
for i in range(N_Pvm):
if(Pvm[i][0]=="flavor3")and(Hdate.month>6):
NEPvm[i]=int(NEPvm[i]*34)
else:
#return
T=0.85
V=6
NEPvm=ES.ES(lenD,N_Pvm,S_Data,Predict_time,T,V)
for i in range(N_Pvm):
if(Pvm[i][0]=="flavor8"):
NEPvm[i]=int(NEPvm[i]*1.7)
# allocation
Servers_N,P_Servers=Allocation.Allocate(NEPvm,Pvm,N_Stype,Servers)
result=utils.results_expression(P_Servers,Servers_N,Servers,Pvm)
return result
def predict_vm(ecs_infor_array, input_file_array):
#Get the CPU information
N_Stype, Servers, N_Pvm, Pvm, Predict_time, Predict_Start = utils.splitEscData(
ecs_infor_array)
#Get the History Data information
length, Hvm, History_End = utils.splitInputData(input_file_array)
#History Data
#Statistic and Split
lenD, S_Data = utils.Statistic_Split(length, Hvm, N_Pvm, Pvm)
# deal with gaps
gaps = int(round((utils.getTimeDiff(Predict_Start, History_End)) / 86400))
for i in range(N_Pvm):
for j in range(gaps):
S_Data[i].append((S_Data[i][-7]))
result = []
if ecs_infor_array is None:
print 'ecs information is none'
return result
if input_file_array is None:
print 'input file information is none'
return result
if (N_Pvm > 5):
T = 2.6
V = 0
#return
NEPvm = ES.ES(lenD, N_Pvm, S_Data, Predict_time, T, V)
else:
T = 0.85
V = 5
NEPvm = ES.ES(lenD, N_Pvm, S_Data, Predict_time, T, V)
# allocation
Servers_N, P_Servers = Allocation.Allocate(NEPvm, Pvm, N_Stype, Servers)
result = utils.results_expression(P_Servers, Servers_N, Servers, Pvm)
return result
def predict_vm(ecs_infor_array,input_file_array):
#Get the CPU information
CPU_kernal,CPU_memory,N_Pvm,condition,Pvm,Predict_time,Predict_Start=utils.splitEscData(ecs_infor_array)
#Get the History Data information
length,Hvm,History_Start=utils.splitInputData(input_file_array)
#History Data
#Statistic and Split
lenD,S_Data=utils.Statistic_Split(length,Hvm,N_Pvm,Pvm)
#print S_Data
result = []
if ecs_infor_array is None:
print 'ecs information is none'
return result
if input_file_array is None:
print 'input file information is none'
return result
#--------------------method five estimate--------------
#NEPvm=TDEstimation.EstTD(History_Start,Predict_Start,lenD,N_Pvm,S_Data,Predict_time)
#-----------------------Mirror-------------------------
NEPvm=mirror.Mirror(lenD,N_Pvm,S_Data,Predict_time)
print NEPvm
#---------------------method Three--------------------
#NEPvm=lwlr.lwlr_predict(N_Pvm,lenD,S_Data,Predict_time)
#print NEPvm
# allocation
CPU,N_PCPU=Boxing(NEPvm,Pvm,N_Pvm,CPU_kernal,CPU_memory)
print N_PCPU
result=utils.results_expression(CPU,N_PCPU,N_Pvm,Pvm)
return result